ImgCoT：将长链推理压缩为紧凑的视觉标记以实现大语言模型的高效推理
ImgCoT: Compressing Long Chain of Thought into Compact Visual Tokens for Efficient Reasoning of Large Language Model
AI总结：
ImgCoT通过将长链推理压缩为视觉标记，提升大语言模型的推理效率，同时保留全局结构和细节信息。
摘要：
将长链推理（CoT）压缩为紧凑的潜在标记对于大语言模型（LLMs）的高效推理至关重要。最近的研究通过自编码器实现这一目标，通过从潜在标记中重建文本CoT来实现，从而编码CoT语义。然而，将文本CoT作为重建目标迫使潜在标记保留表层语言特征（例如词汇选择和语法），引入了强烈的语言归纳偏置，优先考虑语言形式而非推理结构，并限制了逻辑抽象。因此，我们提出了ImgCoT，将重建目标从文本CoT替换为通过将CoT渲染为图像获得的视觉CoT。这将语言偏置替换为空间归纳偏置，即倾向于建模推理步骤的空间布局在视觉CoT中，使潜在标记能够更好地捕捉全局推理结构。此外，尽管视觉潜在标记编码了抽象推理结构，但它们可能会模糊推理细节。因此，我们提出了松散的ImgCoT，一种混合推理，通过在视觉潜在标记中补充少量关键文本推理步骤来增强，这些步骤基于低token对数似然选择。这种设计使LLMs能够在比完整CoT更少的token中保留全局推理结构和细粒度推理细节。在多个数据集和LLMs上的广泛实验证明了ImgCoT两种版本的有效性。
英文摘要：
Compressing long chains of thought (CoT) into compact latent tokens is crucial for efficient reasoning with large language models (LLMs). Recent studies employ autoencoders to achieve this by reconstructing textual CoT from latent tokens, thus encoding CoT semantics. However, treating textual CoT as the reconstruction target forces latent tokens to preserve surface-level linguistic features (e.g., word choice and syntax), introducing a strong linguistic inductive bias that prioritizes linguistic form over reasoning structure and limits logical abstraction. Thus, we propose ImgCoT that replaces the reconstruction target from textual CoT to the visual CoT obtained by rendering CoT into images. This substitutes linguistic bias with spatial inductive bias, i.e., a tendency to model spatial layouts of the reasoning steps in visual CoT, enabling latent tokens to better capture global reasoning structure. Moreover, although visual latent tokens encode abstract reasoning structure, they may blur reasoning details. We thus propose a loose ImgCoT, a hybrid reasoning that augments visual latent tokens with a few key textual reasoning steps, selected based on low token log-likelihood. This design allows LLMs to retain both global reasoning structure and fine-grained reasoning details with fewer tokens than the complete CoT. Extensive experiments across multiple datasets and LLMs demonstrate the effectiveness of the two versions of ImgCoT.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
